Diagnostic performance of a stepwise cytological algorithm for biliary malignancy in primary sclerosing cholangitis

BACKGROUND AND AIMS: Detection of early cholangiocarcinoma (CCA) in primary sclerosing cholangitis (PSC) is challenging. The aim of this study was to evaluate the diagnostic accuracy of a stepwise approach to biliary brush cytology with sequential use of fluorescence in-situ hybridization (FISH) for the detection of biliary malignancy in PSC. METHOD: We retrospectively studied consecutive patients with PSC who underwent biliary brushings at Karolinska University Hospital between 2009 and 2015 (n = 208). Brush samples were categorized as benign, equivocal (atypical or suspicious) and malignant. Equivocal cases were further analysed with FISH. Samples with a malignant cytology or positive FISH were considered positive. The diagnosis was determined after 12 months of follow-up. RESULTS: The diagnosis CCA was confirmed in 15 patients (7%), high-grade dysplasia in three patients, and low-grade dysplasia in five patients at follow-up. Using the diagnostic algorithm, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) for a diagnosis of CCA were 80% (95%CI 52%-96%), 96% (95%CI 92%-98%), 60% (95%CI 36%-81%) and 98% (95% CI 95%-100%). In patients with equivocal cytology (n = 61), the sensitivity for CCA diagnosis increased to 100% (95%CI 72%-100%) with a lower PPV of 58% (95%CI 34%-78%). The diagnostic accuracy for detection of CCA in all patients was 95% (95%CI 91%-97%). CONCLUSION: Biliary brush cytology with sequential use of FISH in equivocal cases seems to be a highly predictive diagnostic test for CCA in PSC. These results support the use of FISH when cytology is equivocal for detection of biliary malignancy in PSC.